Association of arbitrary prescribing behavior to costly drug expenditures: a pharmacoeconomic study in primary care

Background: We aimed to examine the cost-related prescribing performance of primary care physicians who had a higher versus lower tendency of arbitrary prescribing. Methods: In this cross-sectional study, we evaluated the prescriptions of primary care physicians in Istanbul, collected with 3:1 systematic sampling. We determined higher versus lower arbitrary prescribing by the physician’s degree of writing the solo diagnosis of “Z00- General examination without diagnosis/complaint”: those for whom such prescriptions constituted >5% were classified as Group A and those with them at <0.5% as Group B. We compared these two groups by the total and disease-specific drug costs per prescription they produced for 10 frequently encountered indications. Results: The median cost of disease-specific medication for all diagnoses in Group A and Group B, except anemia, was equal. In upper respiratory tract infections (URTIs), hypertension, anemia, diabetes, and conjunctivitis, the mean prescription costs of Group A were significantly higher than those of Group B (P < .001, P < .001, P = .009, P = .007, and P < .001, respectively), whereas disease-specific drug costs per prescription were similar (P > .05 in all diagnoses). In myalgia, Group A had lower cost per prescription (P < .001) and higher analgesic costs per prescription (P < .001) compared to those in Group B. We found significantly higher disease-specific drug cost share in Group B for URTIs (antibiotic), gastroesophageal ref lux disease (gastric acid–suppressant), hypertension (antihypertensive), anemia (iron preparations), diabetes (antidiabetic), depression (antidepressant), and conjunctivitis (corticosteroid) than those in Group A (P < .001 for each). Conclusions: Our study showed that physicians who had a higher tendency of prescribing for no clear indication are also more likely to produce costly prescriptions.